■^^'.h^- TV>ovr»4»S, **!f\ ^ rus. ,V T, ^^ano. >4w^- Class_fe2i Book ^ ^ So f 1 88 1.} The Manuscript Troano. 625 (^From the American Naturalist, August, 1881.) THE MANUSCRIPT TROANO.^ V BY PROFESSOR CYRUS^THOMAS. This manuscript was found about the year 1865 at Madrid, Spain, by the Abbe Brasseur de Bourbourg while on a visit to the Library of the Royal Historical Academy and named by him " Manuscript Troano," in honor of its possessor Don Juan de Tro y Ortolano. So far as I am aware nothing more is known in reference to its history ; we are not even informed by its last owner where or how he obtained it. In ordinary cases this would be sufficient to arouse our suspicions as to its genuineness, but in this case the work itself will dispel all such suspicions. This work was reproduced in fac-simile by a chromo-lithographic process, by the Commission Scientifiqiie du Mexique under the auspices of the French Government, Brasseur de Bourbourg being the editor. The original is written on a strip of Maguey paper about four- ^ Extracts from a paper now being prepared by Professor Thomas for the Bureau of Ethnology, Smithsonian Institution. c 626 The Manuscript Troano. [August, teen feet long and nine inches wide, the surface of which is covered with a white paint or varnish on which the characters and figures are painted in black, red, blue and brown. It is folded fan-like into thirty-five folds, presenting when the folds are pressed together the appearance of an ordinary octavo volume. The hieroglyphics and figures cover both sides of the paper comprising seventy pages, the writing and painting of the figures having been apparently executed, after the paper was folded, so that the fold- ing does not interfere with the writing. • A slight examination of this manuscript is sufficient to con- vince any one at all familiar with Landa's characters that those used here are substantially the same, be the significations what they may. On almost every page are to be found columns of characters agreeing precisely with those given by him as repre- renting the Maya days. Are they used on account of the signifi- cation of the words they represent, as Brasseur supposed, or simply to designate days ? The determination of this point must be one important step toward ascertaining the object and contents of the work. Another prominent feature of the manuscript is the great num- ber of numerals or numeral characters — short straight lines and dots — found on every plate. These, together with the columns of day characters, constitute fully one half the written portion of the work ; hence if we can ascertain the method in which, and the object for which, these were used, sufficient will have been learned to indicate, beyond doubt, the character of the work, and will render the task of deciphering the hieroglyphs much easier than to work at them blindly. Assuming that the reader is familiar with what has already been written upon this subject, I will at once proceed with what I believe to be the correct explanation of the use of these two classes of characters in this manuscript, and which I believe is the key that will ultimately unlock its mysteries. As I shall have occasion to refer very frequently to the Maya calendar, and cannot, without occupying too much space, give here a full explanation of it, I refer the reader to the following easily accessible works : " Bancroft's Native Races," Vol. 11, and Dr. Valentini's article in the Proceedings American Antiquarian Society, giving here only the following brief summary : 1 88 1.] The Manuscript Troano. 627 TABLE I. No. of Mos. Maya Days. I Kan 2 Chicchan .3 ^ Cimi 4 ' Manik s Lamat 6 Muluc 7 Oc 8 Chuen P Eb 10 Ben II Ix 12 Men 13 Cib I Caban 2 Ezanab 1 Cauac 4 Ahau ■; Ymix 6 Ik 7 Akbal Their year consisted of eighteen months of twenty days each, and five intercalated or added days at the end. These added days — to make the full number, 365 — were not counted in any of the months, as the month never counted more or less than twenty days. The names of these twenty days are given in the annexed table. Although they were sometimes numbered from i to 20, yet the usual method, especially in computations of time relating to religious feasts and ceremonies, was as shown in the table. Commencing with i they were numbered to 13, the following day instead of being 14 w^as numbered i, the next 2, and so on to 13. As will be seen from the table, supposing it to represent the first month, the second month would begin with 8 Kan, and so on through the year as shown in the following table of the months and days : TABLE II. E ^ Numbers of the Months Names of Days. Kan , Chicchan Cimi , Manik Lamat , Muluc Oc Ciiuen , Eb , Ben , Ix Men , Cib Caban Ezanab , Cauac , Ahau , Ymix , Ik Akbal U *3 O IE" 13 14 IS 16 17 18 13 Kan Chicchan Cimi . Manik. . . Lamat . . . 628 The Manuscript Troano. [August, If the first day of the year was Kan, as in this table, then each month would commence with Kan and end with Akbal, though numbered differently. If the last day of the i8th month was 9 Akbal, as shown in this table, the five added days would be 10 Kan, 1 1 Chicchan, 1 2 Cimi, 1 3 Manik and i Lamat ; the first day of the next year would be 2 Muluc. When the year began with 2 Muluc, the last day of the iSth month would be 10 Lamat and the five added days would be 1 1 Muluc, 12 Oc, 13 Chuen, i Eb and 2 Ben. The next year would then begin with 3 Ix. Following out this process we shall find the years commencing as follows: i Kan, 2 Muluc, 3 Ix, 4 Cauac, 5 Kan, 6 Muluc, 7 Ix, 8 Cauac, 9 Kan, 10 Muluc, 11 Ix, 12 Cauac, 13 Kan, i Muluc, 2 Ix and so on, the first day being in all cases one of these four. As 13 is a prime number it will require a cycle of 52 years — 13 x 4 — before we again reach i Kan. I give here a table of one of these cycles, showing the order of the years for this length of time. The names by which the years of the different columns are designated are given at the head of the columns. Although their system was somewhat compli- cated by this singular method of numbering the days and years, still it is not difficult to understand it so far. But in order to further complicate this calendar, which was undoubtedly devised by the priests as Landa truly says, " to deceive this simple people," another period called the Katun or Ahau was introduced. This period, according to most authori- ties, consisted of twenty years, but according to Perez of twenty- four. Instead of being numbered in regular order, one, two, three, &c., these periods were also numbered by the thirteen series, but in the following singular order: 13, 11, 9, 7, 5, 3, i, 12, 10, 8, 6, 4, 2, the 13th Katun preceding the nth, and so on. The chief difficulty experienced in attempting to bring this period into harmony with the system so far as given, is, ist. The uncertainty as to whether it consisted of 20 or 24 years ; 2d. To place these periods in their proper positions in the great cycle, that is, to determine what year in any cycle was the first year of a Katun. If this can be done, then it is not difficult to compare TABLE III. cj 6 ^ c3 ■3 3 6 I 2 3 4 5 6 7 8 9 10 II 12 13 I 2 3 4 5 6 7 8 9 10 II 12 13 I 2 3 4 5 6 7 8 9 10 II 12 13 I 2 3 4 5 6 7 8 9 10 II 12 13 i88i.] The Manuscript Tr^ano, 629 the years of the Maya calendar with those dated from the Chris- tian era, if any one can be determined. The reader will observe that we have so far proceeded upon the assumption that the first year of the cycle was i Kan, or in other words, that the Kan column always occupied the extreme left. That the four days, Kan, Muluc, Ix, Cauac — or " year bear- ers," as they were called by the Mayas — must follow in the order given, is manifest, but that Kan must come first does not follow from anything apparent in the system itself; either day may be the first, without any change in the system, but not without a differ- ence in the result. There are some reasons apparent in the manu- script itself for believing that the author considered Cauac the first, or ruling day, and hence Kan the second, Muluc the third and Ix the fourth. One of these reasons will be given here- a,fter. The importance of knowing which one of these days came first will be apparent from the following illustration: A certain event, for example, is dated a particular day in the year, i Ix. By reference to the tables we give here — one commenc- ing with Kan and the other with Cauac — we see that if the forro^er be the correct one, the year i Ix would be the 27th year of the cycle (it is marked with a star) ; if the latter, it would be the 40th, or 13 years later. As we shall have to re- fer very frequently to the calendar, it becomes neces- sary that we construct one. Since the system admits of fifty-two changes in the day on which the year begins, it would require fifty-two calendars to include the years of one cycle, just as four- teen are required to suit all the years of our system — seven for the ordinary and seven for the leap years. TABLE IV. TABLE V • d u 0 d _S csi rt 3 i "3 I— 1 s rt 0 § rt 0 I 2 3 4 I 2 3 4 5 6 7 8 5 6 7 8 9 10 II 12 9 10 II 12 13 I 2 3 13 I 2 3 4 5 6 7 4 5 6 7 8 9 10 II 8 9 10 II 12 13 I* 2 12 13 I 2 3 4 5 6 3 4 5 6 7 8 9 10 7 8 9 10 . II 12 13 I II 12 13 I* 2 3 4 5 2 3 4 5 6 7 8 9 6 7 8 9 10 II 12 13 10 II 12 13 As it would require much time and space to write these out in 630 The Manuscript Troano. [August, full, I have adopted the expedient shown in the following table of abbreviating the work : TABLE VI. < ^•5 5.0 0 t-t N fO •>!l- 10 VO t^ 00 ON 0 ro M 10 VO t^ 00 ON 0 ro t^ 00 0\ 0 l-t C^ t-i HH N ro ^ 10 VO r^ 00 ON 0 M N ro M <^ M N ro n- 10 VO r^ 00 ON 0 Kl ro M M r<-j Ti- ui VO ^ VO t^ 00 On 0 1^ fO M N ro "* 10 VO t^ 00 On 0 )-4 « N M 0 t-4 N m Tl- XO VO t^ 00 ON 0 hH ro M « N ro ^ m a\ m 0 t^ 00 ON 0 « M ro " N ro "* vr> VO Jr^ 00 On 0 M 00 M N fO " N to ^1- U-) VO t^ 00 On 0 ro « N ro rt- t^ rt- 10 vO t^ 00 0\ 0 M N H4 ro M N ro -* in VO t^ 00 On 0 vo 0 M N « c« to Th u-i VO t^ 00 ON 0 M N ro M C^ CO VO 00 ro -* u-> vO r^ 00 ON 0 ro t-i M N ro "^ IT) VO t^ 00 ON Os 0 M ro M N CO 'I- 10 VO t^ 00 ON 0 hH 1-4 ro M M N M N ro rh VO vO r^ 00 ON 0 i-i 11 ro M « ro 't U-) VO r^ 00 CO On 0 t-( N « N ro Tj- \ri VO t^ 00 On 0 hH w 1-H ro IH 1^ »-i N ro ■!t ii-> VO i^ 00 ON 0 i-t ro « N CO T^ 10 VO t^ t?; w nf Q Is 0 0 _3 u hH 4) J3 0 ,0 U (Si l-H a! ,0 < G "S. s 0 0 0 3 (0 eq <1 fl (fl 0 3 "3 c S 3 U 1 U 0 ^ w 1— 1 (3 G ni CJ 0 oi =1 12, (?;, 7, (?), 2, 6 — two of them being obliterated. 634 The Manuscript Troano. [August, Turning to Plate xxii we find the character for Kan repeated in the same way with the numerals over them as follows : ii, 2, 6, lo, I, 5, 9. (?). 4, 8, 12, 3, 7, II. On Plate xxi the character for Muluc is similarly repeated, and the numerals are I2, 3, 6, 10, i, 5, 9, 13, 4, 8, 2, 7, 3. On Plate xx Ix is similarly repeated, and here the numerals are 13, (?), 8, 12,3,7,11,2,6, 10,1,5,9. If we construct a table of years for two cycles, and compare these numbers with those in the table, we shall find the two pre- cisely the same and in the same order, except the Muluc column which presents a partial variation which I will endeavor pres- ently to explain. We give both Cauac and Kan tables. TABLE VII. TABLE VIII. If we run our eyes down the Cauac column of either table until we reach 10, we shall find the numbers from thence down- wards as follows : 10, i, 5) 9> 13,4, 8, 12, 3, 7, II, 2, 6, pre- cisely as in Plate xxiii. The same thing is true in reference to the Kan and Ix columns. The numbers on Plate xxi (Muluc) after the first two — 12 and 3 — skip to 6 and continue regularly from thence to 8. If we start with 12, we find the next to be 3 as given, the next is 6 instead of 7, as it should be ; we notice that in the ad- joining Kan column the next figure is 6, and the numbers thence to 8 as given. May we not suppose that the author of the manuscript had a similar table (with numeral characters) before him, and that in copying 13 I his eye fell on the wrong col- 4 5 umn? That such tables were used by them is rendered prob- able by the following quotation U! % I 2 3 5 6 7 9 10 II 13 I 2 4 5 6 8 9 10 12 13 1 3 4 5 7 8 9 II 12 13 2 3 4 6 7 8 10 II 12 12 3 7 II 2 6 10 I 5 9 13 4 7 iij 12 3 9 10 I 13 4 5 8 9 12 13 3 4 7 8 II 12 2 3 6 7 10 II I 2 5 6 13 q|io I 5 9 13 which Perez makes from an ancient manuscript in his possession : 1 88 1.] The Mannsaipt Troano. 635 " There was another number which they called Ua Katun and which served them as a key to find the Katunes ; according to the order of its march it falls on the Uayeb haab, and revolves to the end of certain years, Katunes 13, 9, 5, i, 10, 6, 2, 1 1, 7, 3, 12, 8,4." By commencing at the bottom of either column of our table of years, and running up, we will find precisely these numbers, and in the order here given. It is scarcely possible these could have been obtained except by a table of years similar to those we have given. Be this as it may, the fact that these numerals and char- acters, as here interpreted, include a continuous period, is too plain to be ignored. The agreement in so many numbers and the order in which they come cannot be accidental. From this we are justified in concluding, ist. That these day characters are used simply to represent days ; 2d. That the red numerals are used to denote the days or years of the Maya "week " (as their period of 13 days and years has been termed, though they applied no name to it), which is corroborated by the fact that with the exception of two on the title page, none in the manuscript denote a greater number than 13 (there is one other apparent exception, but the additional dot is a blotch or evident mis- take); 3d. That the day columns are to be read from the top down- wards; 4th. That Landa's characters for the Maya days are correct; and 5th. That the work is some kind of a calendar, probably, con- taining directions to be followed by the priests and people in refer- ence to their religious duties. Now let us apply our theory to the day columns and numerals found in some of the other plates. We select as our first example the column in the lower division of Plate XXVI, as here both the red and black numerals are 1 3 throughout. The day characters are those for Ahau, Eb, Kan, Cib and Lamat, in the order here given. According to the inter- pretation suggested, the red numerals refer to the days of the week and the black to the days of the month. Proceeding upon this assumption, we will now try to find out, by using these numbers, in what years 1 3 Ahau, 1 3 Eb, 1 3 Kan, 1 3 Cib and 1 3 Lamat are to be found in the 13th month. The result is as follows : Years . 13 Ahau. ijEd. ij Kan. 13 Cib. 7j Lamat . 6 Cauac 7 Cauac 2 Cauac 3 Cauac II Cauac . 4 Kan 12 Kan 7 Kan 8 Kan 3 Kan . 9 Muluc 4 Muluc 5 Muluc 13 Muluc I Muluc . I Ix ' 2lx 10 Ix Six 6Ix TABLE IX. The Manuscript Troano. [August, o OS 0 .1— ( lo II 12 13 I 2 3 4 5 6 7 8 9 10 II 12 13 I 2 3 4 S 6 7 8 9 10 11 12 ^ "T 2 3 4 5 6 7 8 9 10 II 12 13 ^ II 2 6 3^ 4 ^ __7, 1^ 9 10 II ' xz "'13 \ p-. 2 Z i 5 6 ^7 1 ^9 10 II 12 .3] 1^3 I 2 u S R-7 II ^8f9' 10 II In order to bring clearly before the eye the places in the cycle where these years fall, we give here the Cauac table governing the period embraced in the four plates heretofore alluded to. Marking the numbers, we find a continu- ous period of twenty years, possibly one Ka- tun, though not in the precise order we would expect it, that is, it does not include five com- plete transverse lines. If we use the Kan table, we obtain the same result, except that then we shall have three years in the uppermost line and but one in the lowest. The years of this period are surrounded by a continuous dark line. As Plate XXVII appears to be a continuation of the same subject as that presented on Plate XXVI, we select the day column of the upper division. The days are 11 Ahau, 11 Eb, 11 Kan, 1 1 Cib and 1 1 Lamat — the months all the 1 3th. This gives us the following years : // Ahau. Years .... 4 Cauac " .... 2 Kan " .... 7'Muluc , 12 Ix ij Eb. 5 Cauac 10 Kan 2 Muluc 13 Ix II Kan. 13 Cauac 5 Kan 3 Muluc 8Ix . II Cib. I Cauac 6 Kan II Muluc 3 Ix 11 Lamat. 9 Cauac I Kan 12 Muluc 4 Ix The numbers which are surrounded in the table by a waved line, also form, as we see, a continuous period of twenty years. We likewise observe that between the periods there is an interval of four years. Plates VII, VIII, IX and x, of the second part of the manuscript, furnish perhaps the strongest proof of the correctness of my in- terpretation of the red numerals. The middle division of these plates evidently refers to one subject. Here we find thirteen short columns, of three day characters each, inserted in the text, each character with red numerals over it, and each column with black numerals at the bottom denoting 17, as the month. We give here the names of the days with the number of each as shown by the red numerals ; the order in which the columns occur is i88i.] The Manuscript Troano. 637 6Cib 13 Cib 7 Caban I Caban 8 Ezanab 2 Ezanab 9 Cib 3 Cib 10 Caban 4 Caban II Ezanab 5 Ezanab 5 Cib 12 Cib 6 Caban 13 Caban 7 Ezanab i Ezanab 1 Cib 8 Cib 2 Caban 9 Caban 3 Ezanab 10 Ezanab 2 Cib 3 Caban 4 Ezanab also preserved although we have strong doubts as to the correct- ness of Brasseur's paging. 4 Cib II Cib 5 Caban 12 Caban 6 Ezanab 13 Ezanab 10 Cib 7 Cib 11 Caban 8 Caban 12 Ezanab 9 Ezanab We see by examining the list of days in the Table No. 11, that these three days follow each other in the order here given, thus : if the first is 6 Cib, the next is 7 Caban and the next 8 Ezanab. It follows, therefore, that the three days of any one of these groups must fall in the same month^ and year, hence we have to search for but four years for each column — but 4 x 13 = 5^ years, an entire cycle. As the three numbers in a group will sufficiently designate the group, we will omit the names. Years. Years. Years. Years. TABLE X. 6, 7, 8 7 Cauac 13, I, 2....- I " 4. 5, 6 5 " II, 12, 13 12 " 5, 6, 7 6 « 12,13, 1 13 " 2, 3. 4 3 " 9, 10, II 10 " 3. 4. 5 4 " 10, II, 12 II " 7, 8, 9. 8 " I, 2, 3 2 " 8, 9, 10 9 " If we mark in the table the numbers cor- responding with these years, we find that they make one complete cycle, neither more nor less. It is true we should find the same result, no matter where we begin in the cycle, but the point in- sisted on is, that they form a continuous term corresponding with one of the Maya periods. We give one more example. In the second division of Plates xxx and XXXI commencing on the left half of the former and continuing through the latter, we find a series of similar figures, except the -It is not necessary as a matter of course, that three successive days always fall in the same month, but in this case they do. In the Dresden codex plates 51-58, we find similar three day columns, some of which do not conform to this rule. Years. Years. Years. 12 Kan 4 Muluc 9 Ix 6 a n " 3 " 10 (( 2 « 7 " 4 " 9 " I " II ee 3 " 8 " 5 « 10 " 2 « 8 « 13 '* 5 " 2 tt 7 " 12 " 9 te I " 6 « 3 " 8 " 13 " 13 << 5 " 10 " •7 (C 12 " 4 " I ti 6 " II " 0 0 6 10 II 12 13 I* 2* 6* 3* 7* 4* 8* Q* 10* II*, 12* n* I* 2* 1* 4* 8* 5* 6* 10* 1* II* 12* IS* I* 2* 3* 7* 4* 8* 5* 9* 6* 10* n* 12* IS* I* 2* 6* 3* 7* 4* 8* 5* Q* 10* II* 12* 13* I 2 3 4 638 The Manuscript Troano. [August, one on the right of the first plate, which is the god Tlaloc, or Maya equivalent. Over each figure are red numerals differing in number, and in front black numerals all denoting 11. The red numerals are (?) 9, 7, 5, 3. The first is obliterated, but judging from the space, is i, but by the succession, is 1 1 ; however, as the result will be the same except as to the position of the period in the table, it does not make any material difference for present purposes, which we select. For reasons not necessary to be dis- cussed here, notwithstanding the strong evidence afforded by the succession of numbers, we are inclined to believe the missing number was i, and that the order was 9, i, 3, 5, 7. The days found at the left of the compartment on Plate xxxi, are Kan, Cib, Lamat, Ahau, Eb ; as each numeral applies to each day, the number of years indicated will be 5 x 5 X 4 = ioo> ^"^^ will be as follows, the month being the i ith in each case : 9 Kan. 9 Cib. g Lamat. 9 Ahau. gEb. Years.. ..12 Cauac 13 Cauac 8 Cauac 3 Cauac 4 Cauac <( . . 4 Kan 5 Kan 13 Kan I Kan 9 Kan " .. . . 2 Muluc 10 Muluc II Muluc 6 Muluc I Muluc « .. 7lx 2 Ix 3 Ix II Ix 12 Ix 7 Kan. 7 Cib. 7 Lamat. 7 Ahau. 7 Eb. Years . . . . 10 Cauac II Cauac 6 Cauac I Cauac 2 Cauac « . . 2 Kan 3 Kan II Kan 12 Kan 7 Kan « ..13 Muluc 8 Muluc 9 Muluc 4 Muluc 12 Muluc •< .. Six 13 Ix I Ix 9 Ix 10 Ix 5 Kan. 5 Cib. 5 Lamat. J" Ahau. SEb. Years.. . . 8 Cauac 9 Cauac 4 Cauac 12 Cauac 13 Cauac " .. ..13 Kan I Kan 9 Kan 10 Kan 5 Kan <( ..II Muluc 6 Muluc 7 Muluc 2 Muluc 10 Muluc " .. . . 3 Ix II Ix 12 Ix 7IX 8Ix 3 Kan. 3 Cib. 3 Lamat. 3 Ahau. 3Eb. Years . . . . 6 Cauac 7 Cauac 2 Cauac 10 Cauac II Cauac « .. ..II Kan 12 Kan 7 Kan 8 Kan 3 Kan (< . . 9 Muluc 4 Muluc 5 Muluc 13 Muluc 8 Muluc <( .. I Ix 9IX 10 Ix Six 6Ix I Kan. / Cib. / Lamat. I Ahau. I Eb. Years . . . . 4 Cauac 5 Cauac 13 Cauac 8 Cauac 9 Cauac " ,. . . 9 Kan 10 Kan 5 Kan 6 Kan I Kan (( . . 7 Muluc 2 Muluc 3 Muluc 11 Muluc 6 Muluc " .. . .12 Ix 7lx Six 3IX 4lx We give here a table running through four cycles, on which we mark the five periods indicated by the preceding list. Each, i88i.] The Manuscript Troano. 639 13* 4* 12* 3* \A 10 flT 2* 6* 10* I* 12* 3* 7* II* 7 8 9* 13* 10* I* II* 12* 2* t 4* 5* 6* 7* 8* 9* 10* I* II* 12* 13* 2* I3 4 as a. matter of course, embraces twenty years, but we observe TABLE XI. between each period four uncounted years. We must not presume from this that these years are never counted, as this is contradicted not only by the pre- ceding example but by others which might be presented ; still it does indicate that the view maintained by Perez was correct. Assuming that a Katun consisted of twenty-four years, we can see from this table why they were numbered in the pecu- liar manner heretofore mentioned. We see that the years with which the five, here marked, begin, are 11, 9> 7> 5> 3> i'^ other words this order necessarily follows if the Katun or Ahau (we prefer the name Katun) consists of 24 years. If the upper and lower dark lines are extended across so as to include the whole 24 years, the periods will then commence with the years 9, 7, 5, 3, i, and will, as I believe, give us the correct numbers of these Ahaues. Our space will not permit us to give further examples or to discuss further the meaning and use of the numerals. But before closing we would ask the reader to turn again to our table No. vii. showing the period indicated by the characters on Plates xx — xxiii. As we see, this embraces part of two cycles, and includes parts of at least three Katuns. If I am correct, it will then include a period of time which was considered notable by that people. If 5 ,i. 7* II* 8* 12* 9* 10* 13* I* 2* 3* 4* 5* 6* 7* 8* 9* 13* 10* II* 12* I 2 3 4 5* 9* 6* 10* ri" -w II* 12* 13* I* 2* 3* 4* 5* 6* 7* II*] 8* 9* 10* 12 13 I 2 3* 7* 4* 8* 5* b* 9* 10* II* 12* 13* I* 2* 3* 4* 8* 5* 9* 6* 7* 10 I II 12 13 2 3 4 5 6 7 8 9 10 II 12 13 I 2 3 4 5 6 7 8 9 10 II 12 13 I 2 3 4 5 6 7 8 9 10 II 12 13 I 2 3 4 5 6 7 8 9 10 II 12 13 I 2 3 4 5 6 7 8 9 10 II 12 13 I 2 3 4 S 6 7 8 9 10 II 12 13 640 The ManiLScript Troano. [August, we turn to Plate xxxiii, where Cauac is the ruhng character, we find, as I think, in the figures of the upper compartment, and in its ground-color, which is of a brownish or twilight tint, indica- tions of the close of some important period of time. We see Ahau with a fiery red face sinking out of sight in the lower, left- hand corner, casting back, as represented by the torch in his hand, his fiery beams as he sinks below the western horizon. We say " sinks below," because, just above is another (headless) figure on which is the character Caban or Cab which signifies " to descend," and the machete. The dark figure to the right drops ten white balls, to the last or upper of which he points one finger; while just above it we see a single dark one, with a slender streak of light in it, just leaving the hand. If we cast our eyes over the Cauac column of the tables we see that ten, of the first cycle, is followed by one, of the second. This 10 corresponds with the 10 Cauac year in the next to the last group in Table xi. To the left, and directly in front of the face of the dark figure, are two characters connected, the lower one Cauac, the upper a rather unusual form of Ix. We have in this a symbol of the meeting of the two periods, the one ending with 13 Ix the other commencing with I Cauac. The most conspicuous object in the compartment is the bright blue figure in the upper left-hand corner, on a black ground, surrounded by a white bordering which is crossed by dotted rays, each ending with a small round character. By counting, we find there are just twenty-three of these rays, but exactly where the dagger crosses to the eye, is a space for one more — the 24th. One year more will complete the Ahau and then its light will go out forever. If we notice the single red dot over the second Canac character in the left-hand column of this Plate (xxiii), we observe that it is surrounded by a circle of dots, indicating its importance in some respect. We might go on to show that the figures in the upper compart- ments of the four Plates xx — xxiii all relate to the close or com- mencement of some important period, but our limits will not permit of this, moreover our only object in presenting the fore- going is to call attention to the evidence of the following con- clusions : 1st, That the work is a religious calendar of some kind; 2d, 1 88 1.] The Manuscript Troano. 641 that the day characters are used for the purpose of designating the days and not for the signification of the words ; 3d, that Landa's characters for the day are at least substantially correct; 4th, that the columns composed of these characters are to be read from the top downwards; 5th, that the red numerals refer to the days or years of the week ; and 6th, that the black numerals in connec- tion with the day characters, as well as those in the spaces, usually refer to the numbers of the month. There are some apparent exceptions, but I have ascertained, as I can show, that these are only apparent and not real exceptions. I may also add that we have in what has been shown strong evidence that Perez was right in asserting that the Ahau consisted of 24 years. I have succeeded in deciphering a number of the hieroglyphics of the text but will not enter upon an explanation of these at present, as I have only commenced this part of the work. I ..x-5^